of crevassed ice, some 4 miles wide where Marvine Glacier emerges, broadened toward the sea (Fig. 8), being about 5 miles wide opposite Blossom Island, and on the shores of Yakutat Bay extending from the Kwik River at least as far as Point Manby. For a distance of over 15 miles, from the mountains to the seaward face of the Malaspina, the ice was crevassed over a width increasing from about 4 miles near the mountains to 12 or 13 miles near the sea.
Near Blossom Island the senior author found this lobe of the glacier very different from its condition when Russell and others walked and sledged over it. It was im-
Fio. 3. EXTENT or CBHVASSING IN 1906 IN MABYCNE LOBE or MALABPINA GLAOTEIB.
passably crevassed outside its mountain valley. There were many pinnacles and arfites near the margin, and in the middle were flat-topped seracs surrounded by deep crevasses extending in all directions. It was clearly evident that the breaking in the expanded piedmont bulb in 1906 had been far less intense than hi the mountain valley. The broken area had not been long exposed to ablation, for the flat-topped seracs had been so recently cracked that melting had not had time to round off their corners. The fact that not only the snow-covered, but even the bare serac-tops were not notably affected by melting proved that most, if not all, of the breaking had taken place after the period of ablation of 1905 had ceased.
Sledging over this surface, which Russell and others had crossed easily, would have